Some casual thoughts on the iPad 2: An email response.

Today, I responded to an email from a friend asking about a debate that's going on about the next iPad. There have been numerous reports in recent weeks that the next version of Apple's iPad will feature a double-resolution display that takes it from the current 1024 x 768 resolution to a massive 2048 x 1536. The manner in which this would be accomplished is by simply quadrupling the pixels form the current display in the same way that the iPhone 4 doubled the resolution of the previous iPhone from 480 x 320 to 960 x 640. The reason Apple chose to exactly double the resolution was to make app scaling easy and seamless, therefore maintaining compatibility with all of the existing apps in its App Store. It stands to reason that Apple will take the same approach for the iPad, but there are technical and cost considerations that would have to be overcome. John Gruber of DaringFireball.net agrees that Apple will take this approach eventually, but poured cold water on that happening with the new version expected between now and April.

I won't summarize the arguments here. You can read the different predictions on Engadget, AppleInsider, or the above Daring Fireball link. What follows, however, are some thoughts on Apple's approach to product iteration in an email response this morning:

----------------------

Q: I wonder if Apple leaked directly to Gruber to dehype the frenzy over the retina display. Anything less than retina at this point is going to be a huge disappointment.

Gruber just seemed so sure of it. It must be a direct Apple leak.

Also, nobody has directly addressed how much RAM and CPU graphics power this will take. My guess is it is well within the specs for iPad 2. Cost may be an issue this year, but it would be cool to see Apple just go for it. They would be a solid 2-3 years ahead of anybody.

A: Believe it or not, I did wonder the same thing. When Apple gets near product announcements, I usually base my predictions on whether or not they've tried to kill specific hype. So "they" either leaked it on purpose or Gruber has some insane sources. You'll recall, I did doubt the retina display for this year...

As far as processing power and Apple just going for it, I think it matters how you view Apple and how they think. You say that they should just do it because they would be 2-3 years ahead of everyone else, but I think they already think they are. They are so measured in their updates when they don't feel pressured, and this is such a case. Tim Cook indicated as much in the earnings call Tuesday when he said the only obstacle seems to be production and that nobody had a product on the market that is competitive.

Think about it. They could refine the iPad this year with a few tweaks and still sell tens of millions. Then next year when it seems like they need another hook, they release a retina display. That keeps the iPad at the front of the line in 2012, AND enough time has passed that early adopters go ahead and buy a new one. All of these are just strategic decisions completely independent of supply costs and margins, which are so paramount at Apple. If the screens cost a ton, why put them in this year when they could still sell as many as they can make? There's always next year with lower costs.

Also, the reason nobody's addressed CPU and RAM requirements directly is because nobody knows how to. An optimized chip is an unknown with what looks like a Cortex-9 arm processor and some new mobile graphics chip. It'll be dual-core, but not much else can be said. I do have my doubts about it being able to drive that display well, though. The last thing Apple needs is to take a risk on performance when a lesser display works for now.

So... yeah. I hope I'm wrong, though. Kind of difficult putting Engadget and Gruber against each other, since they both are usually very reliable. What I will say is that where Engadget has been wrong in the past is in publishing information a product cycle too early, like when they said there would be apps on the new Apple TV. That could easily be the case here, as Apple will obviously double the resolution at some point.

The iPad is so ubiquitous it's in headlines about Rep. Giffords

It would be a little unnecessary to say that Apple hit it big with the iPad this year, selling 14 million in a new product category in just three quarters.  What's just as remarkable, however, is how ordinary it's become to see references in the broader U.S. culture.  This point is underscored today as the news, and headline in Reuters's case, that Representative Gabrielle Giffords is not only standing with assistance, but is trying to speak, and is using an iPad.  Her doctors caution that she has a long road ahead, and we're sure that everyone wishes her the best in her continued recovery.

It is amazing, though, that doctors felt the need to mention her iPad use.  Any variety of gadgets could have been inserted into this story, but it's an iPad that she's using and they singled out.  It says a lot about Apple's brand that they would mention it, but it's also worth noting that an iPad is probably the best connected device for a patient in stable, but limited condition.  It's hard to imagine a laptop inserted into this story.  Contrary to Apple satire, the iPad may not be able to heal her, but we're sure she's glad something in in its place to keep her connected.

Mobilified Reveal: Android Fragmentation

One of the most overused terms in the mobile space is "fragmentation" when referring to Google's Android operating system.  It has come to refer to a variety of different issues, some bad, some good, that are facing Google as it moves forward as the most widely-used mobile OS.  Fragmentation can refer to the different types experiences that the end-user encounters.  It can also refer to the different challenges that Android presents to developers as they try to create an app that runs well on as many Android devices as possible.  You can think of these two general categories as User Fragmentation and Developer Fragmentation.  While Developer Fragmentation certainly results in and can affect User Fragmentation, they are often separate issues that have their own subsets of problems.  Some of these problems are more visible, and some of them are more problematic.

User Fragmentation

User Fragmentation includes some of the most visible types of fragmentation that phone reviewers are most up in arms about.  The types include, but are not limited to:

1. Manufacturer and carrier skins on top of Android, changing the look, feel, and performance
2. Phones running older versions of Android, including 1.6, 2.1, 2.0, and 2.3

When Google introduced the first Android phone, the G1, they worked closely with HTC to deliver a software experience that they could call their own.  It didn't take long, however, for handset manufacturers to start skinning their devices to try and "enhance" the experience.  In reality, the skinning, as it's called, was an attempt to simply differentiate their phones.  In late 2009, HTC themselves began skinning Android 1.6 on the HTC Hero into what they called "HTC Sense."  Motorola introduced their own version called Motoblur around the same time, despite working closely with Google to introduce the first Android 2.0 device, the Motorola Droid.

When Google unveiled the Nexus One the following January, they billed it as a pure "Google experience" device with no carrier or manufacturer customizations.  It was largely seen as an attempt to cut the carriers out of the process of selling a phone, as Google initially tried to sell it unlocked on their website.  What went unnoticed at the time, however, was that it would be one of the only phones running stock Android released in 2010.  What followed was an Android explosion, one that put Google on the map but also diluted the Android identity.  In addition to HTC's Sense and Motorola's Blur, there was Samsung's Touchwiz, LG's unnamed skin, and the different UI running on Sony's Xperia line.  In short, no major phone maker was releasing Android with the same look and feel.  They all had different homescreen layouts, different menus, different features, and performed with varying degrees of bugginess.  The only other major phones released in 2010 running stock Android were the HTC G2 and the late-in-the-year Nexus S.

Adding to Android's problems was the fact that manufacturers and carriers were doing a terrible job of keeping the phones up to date.  While Android 2.2 "Froyo" was revealed in May and pushed to the Nexus One over the course of months, phones still shipped with 1.6 and 2.1 all the way through the year.  What's worse, consumers were left not knowing if their phones would be updated, including when.  In January 2011, the Samsung Galaxy S line, released last summer, still has not received updates to 2.2.  Many phones barely a year old are forever stuck on Android 2.1 with no support from manufacturers, carriers, or Google to try and bring them to 2.3 "Gingerbread."

These issues are a huge frustrations for a tech community that is well aware of what features are missing or how terrible some of the skins are compared to stock Android.  For average consumers, though, it's not as big an issue.  The typical phone buyer takes what's available from the carrier and isn't aware of the greater Android ecosystem.  This actually hurts the Google brand as much as customers, as people don't think of their phones as Android phones, but rather, "Droids," an "HTC Evo," or a "Samsung Vibrant."  It simply downgrades the overall experience, leaving Apple's iPhone as a recognizable, premium product.  Further hurting the Android ecosystem, though, is what we call "developer fragmentation."

Developer Fragmentation

Developer Fragmentation is the type of fragmentation that is behind the scenes.  It's the stuff that the public doesn't see, but definitely feels in the long-run.  Developer Fragmentation includes all of the variables developers must account for in order to be sure that a particular app will run well on the majority of Android devices.  These issues include:

1. Different screen sizes and resolutions
2. Different external hardware, i.e. screen technology used, touchscreens, camera buttons, trackballs
3. Different hardware internals, particularly processor types, architecture, speed, and amount of RAM
4. Different marketplaces for users to download Android apps

In short, developer fragmentation makes it more resource-intensive to make Android apps.  The reason is that for every different screen size, resolution, and other hardware issue it takes more and more versions to be made to support all of the devices on the market.  It also lessens the quality of the apps, as apps have to appeal to a lowest common denominator in order to run.  Button sizes can't be optimized when you're developing for  both 3.2-inch and 4.3-inch screens.  Graphics can look odd.  Different processor types mean major differences in how apps run, and different hardware technology requires more bug testing to make sure that touch response is not more buggy than usual.

The ways this can impact the final result are dramatic.  For instance, more people working on one app diverts resources from potentially new apps.  For manpower means more risk for developing, which is why Apple's iOS App Store remains dominant in quality app totals.  In addition to raw numbers, the quality of apps suffer because of the lowest common denominator factor mentioned above.  Cross-platform apps are overwhelmingly superior on iOS, despite the fact that focused effort can result in a good app on Android.  One infamous example is what happened to Angry Birds on Android.  Shortly after Angry Birds hit the Android Market, users on some older and low-power phones began experiencing severe performance issues.  The game's maker Rovio apologized for the fragmentation issue on Android and announced it would be releasing a different version that would run better on those phones.  That was in mid-November and the updated version still hasn't arrived.  Rovio also recently discussed how the fragmentation issue affects developers in terms of the different payment systems and growing number of stores for apps.  Amazon recently announced its plans to build an Android app store.  Expect carriers, retailers, and manufacturers to further complicate the ways to purchase Android apps, further frustrating developers and keeping some from fully committing to the platform.

Takeaway

While the word "fragmentation" has almost been reduced to a buzzword in reference to Android, it is a real issue for consumers.  Generally speaking, most Android apps run on most Android phones, but that's not the whole story.  The variety of issues that contribute to developer fragmentation are what can affect a platform more than what greets the consumer visually.  While skins and custom ROMs do complicate the ability to update phones, it's the more base-level issues that affect developers and therefore the Android ecosystem as a whole.  Now that manufacturers, carriers, and other entities have gotten their first taste of making Android fit their needs, the problem is only going to get worse in the years to come.

Sprint raises 3G data rates, losing key differentiator.

Sprint Nextel Corp. announced today that it is raising its 3G data plans by $10.  Sprint's wildly-successful Simply Everything plan has been an asset for their customers as virtually the cheapest all-you-can-eat plan in the U.S.  For $69.99, Sprint customers could get a quality Android phone along with unlimited data, voice, and text.  For an increasingly reliable Sprint network, this was a great deal and was one of the few reasons for people to choose Sprint over some of the larger Verizon and AT&T networks. Sprint began to chip away at this deal last spring with the introduction of the Evo 4G.  Because it was capable of using their 4G WiMAX network, they charged an extra $10 for "premium data services" whether or not you lived in a 4G city.  Since the launch of the Evo, Sprint has not released a single high-end Android phone that didn't use 4G, essentially making the most desirable tier of data a $79.99 plan.  In many ways, Sprint is simplifying (oh, the irony) the issue today by requiring the $79.99 monthly price.  It's seemed for months that they were moving this direction anyway.

Hopefully this won't hurt Sprint in the long run.  For all of AT&T's network issues, Sprint is still considered the third carrier in the U.S. and people will want to know why their plans are edging closer and closer to Verizon and AT&T.  The other issue is that for Sprint to raise these rates and justify it with the same "premium data services" excuse, it isn't likely to make customers happy.   They probably should just call it what it is; a data hike by the carrier that depends upon its identity as a bargain network.

HPretty Encouraging

Engadget has certainly had a great run of exclusives lately, and this first leak of the HP/Palm tablet is no exception. According to Engadget, this leak comes from a trusted source and is of a 9" tablet named or codenamed "Topaz." Engadget says that a 7" tablet, named "Opal," is also in the works and may hit the market later. Other than having a 1.2 Ghz processor, a front-facing camera, and a Micro-USB port, not much else is known about the device, but we're sure to know something after Feb. 9. Of course, that doesn't mean speculation can't run rampant.

For instance, note that the branding on the back has both of the HP and Palm names. In their invite to the Feb.9 event, HP didn't put the Palm name anywhere. Secrecy couldn't have been the issue, since they said directly that it would be a WebOS event. The two images are both obviously renders, but the back displays FCC markings, so perhaps that's what it will actually look like.

And does it ever look good! The front display shows an attractive WebOS that looks pretty inviting to use. After we expressed our disappointment at the small display on the Palm Pre 2, it actually looks as if WebOS has some serious room to breath and roam free! The way the cards and dock look and fit seems very right. In fact, it almost looks too good. While I do think this will be what the tablet looks like, I'm not sold on the back. For one thing, there is no visible name of the device. HP could have kept the name from making it onto early renders, but it seems like it would fit well between the branding and the FCC markings. If that's the case, who knows what else could be different, including the fairly boring molded plastic and curved back. I expect the back to change and accomodate the name of the device. It's also very possible that the Palm name will be gone by the time the device hits.

One thing that shouldn't change is the placement of the quick action bar. They moved the search box below the dock, a welcome move that consolidates most of the navigation of the device at the bottom. It's difficult to tell if the icons on the right of the screen are notifications or part of the menu system. If they used the extra screen real estate to move the menu option to the bottom, sharing the same vertical space as search, that would be a smart move. They're probably just notifications, but moving the menu would have been another element they could have consolidated. As long as the device can differentiate between the dock, the search bar, notifications and the gesture area, it's a pretty clean interface for the device. Here's hoping that HP also has some phones to back up the powerful initiative reflected in this tablet.

Blackbarely: RIM continues to tread water

Boy Genius Report has had a nice run of leaks detailing RIM's Blackberry updates for later this year.  They show an updated Blackberry Torch with much better internals and a new Blackberry Storm, the Storm 3.  The first two models of the Storm were terrible failures for the company, so you have to question why they would resurrect that line.  The Storm 3 is projected to have the following specs:

• 1.2GHz processor
• 3.7-inch 800 x 480 capacitive display
• Quad-band GSM/GPRS/EDGE
• Tri-band UMTS/HSPA
• 5-megapixel camera with 720p video recording
• 8GB eMMC (storage for apps and data), 512MB of RAM

That's actually a pretty strong step up from previous Blackberries.  Given that the relatively new Torch has a better onscreen keyboard than the Storm ever did, the Storm 3 might actually be more usable with these kinds of specs and a better virtual keyboard.  You have to wonder, though, why they're bothering with a 1.2 GHz processor after RIM co-CEO Mike Lazaridis seemed hell-bent on dismissing the "gigahertz" war just recently.  It's actually nice to see competitive hardware put out by RIM.

Oh but wait, that's not the whole story, since these slides give projections for the Torch 2 to be released in "late Q3 2011."  So how about the new Storm??  Oh, September 2011?  Okay, never mind then.  Nobody will care at that point.  Just to remind you, as Chris Ziegler of Engadget puts it, those specs won't impress anyone as we approach Q4 '11.  And he's right.  We're already seeing dual-core phones set to hit Verizon and AT&T in this quarter.  As usual, RIM will be behind the curve (every pun intended) with these phones, making you question whether or not Lazaridis was just blowing smoke when he said we could expect the awesome QNX interface from the Playbook to hit smartphones when dual-core was possible.  On second thought, you're probably not questioning anything, since you already know the answer.